Hypothalamic neurons can sense and respond to changes in glucose concentration, but the functional significance of this glucose-sensing property remains to be determined. As in pancreatic beta cells, glucokinase (GK) constitutes a key component of the hypothalamic glucose-sensing mechanism. Homozygous whole-body ablation of the GK gene produces lethal neonatal diabetes, and heterozygous whole-body ablation of the GK gene causes impaired glucose homeostasis and obese phenotypes, including reduced POMC and increased AgRP gene expression in hypothalamic neurons. Interestingly, homozygous whole-body ablation of the insulin receptor also produces lethal neonatal diabetes, and neuron-specific expression of the insulin receptor partially rescues this phenotype. We have now demonstrated that heterozygous and homozygous ablation of the GK gene specifically in neurons recapitulates the respective whole-body GK knockout phenotypes. On the other hand, other studies have suggested that GK in hypothalamic tanycytes (a type of glial cell) also plays a role in metabolic homeostasis. We therefore propose that different aspects of metabolic homeostasis are dependent on GK expression in POMC neurons, AgRP neurons, or tanycytes. To address this hypothesis, in Specific Aim 1 we propose to ablate GK specifically from POMC neurons by crossing transgenic mice expressing cre-recombinase under control of the POMC promoter (POMC-cre) with mice in which the GK gene is flanked by lox sites. Complementing these studies, we propose to restore GK specifically in POMC neurons in (homozygous or heterozygous) whole-body GK knockout mice, using the "knock-in" strategy by which Okamoto et al. rescued lethal neonatal diabetes by restoring insulin receptors specifically to neurons in whole-body insulin receptor knockout mice. Mice will be maintained on a low-fat or a high-fat diet. Metabolic phenotypes, including leptin sensitivity, glucose and insulin tolerance tests, food intake, body weight, adiposity, metabolic rate, and temperature will be assessed. Mice will be sacrificed, and gene expression will be determined in hypothalamus and other tissues. We propose similar studies in Specific Aims 2 and 3, using AgRP, or GFAP cre-recombinase (expressed in tanycytes) to ablate or restore GK specifically in in AgRP neurons, or to ablate GK in tanyctes. We anticipate that ablation of GK in specific hypothalamic cell types will partially recapitulate specific metabolic impairments produced by whole-body ablation of GK, and that restoration of GK to POMC or AgRP neurons will reverse specific metabolic impairments caused by whole- body GK deficiency